摘要
在基于内容图像检索中,图像的底层视觉特征和高层语义概念之间存在着较大的语义间隔。使用机器学习方法学习图像特征,自动建立图像类的模型成为一种有效的方法。本文提出了一种用支持向量机(SVM)实现自然图像自动语义归类的方法,基于块划分聚类得到特征向量作为SVM训练样本,实现语义分类器。由于参与聚类的是某类图像所有块的特征,提取的特征更能反映某一类图像特征。实验证明这种方法是有效的。
In the approach of content-based image retrieval, there exists a semantic gap between low-level visual features and highlevel concepts. Using machine learning method to learn image features and to automatically construct models for image classes is a promising way. In this paper, support vector machines are trained for nature image classification. We propose an image representation method based on image partition and region clustering. The mapping which (maps an image to its representation does not really depend on that image alone but on the entire collection of images from which the region groups have been built. The experimental results are promising.
出处
《微计算机信息》
北大核心
2006年第05S期210-212,共3页
Control & Automation
基金
山东省中青年科学家奖励基金资助编号:(2001SD521)
关键词
图像划分
特征矢量聚类
支持向量机(SVM)
图像分类
图像检索
image partition
feature vector cluster
support vector maehines(SVM)
image elassilleation
image retrieval